SIMLIN

SIMLIN predicts S-sulphenylation sites in the human proteome to identify cysteine S-hydroxyl (-SOH) post-translational modifications involved in protein regulation and cell signaling.

Key Features:

• Prediction target: Identifies S-sulphenylation (S-hydroxyl, -SOH) sites on cysteine residues across the human proteome.
• Model architecture: Implements a novel hybrid multi-stage neural-network based ensemble-learning model for site prediction.
• Input features: Integrates sequence-derived features and structural characteristics of proteins to inform predictions.
• Performance: Achieved 88.0% prediction accuracy and an AUC of 0.82 on independent testing datasets.
• Benchmarking: Demonstrated superior performance compared to existing state-of-the-art S-sulphenylation predictors in independent evaluations.

Scientific Applications:

• High-throughput in silico screening: Enables large-scale computational identification of potential S-sulphenylation sites for downstream study.
• Hypothesis generation and validation prioritization: Supports selection of candidate cysteines for experimental validation of PTM-mediated regulation and cell signaling roles.
• PTM research: Assists bioinformatics investigations into the role of cysteine oxidation and reversible S-hydroxylation in protein function.

Methodology:

Uses a novel hybrid computational framework comprising a multi-stage neural-network based ensemble-learning model that integrates sequence-derived features and structural characteristics, with performance evaluated by benchmarking on independent testing datasets (88.0% accuracy, AUC 0.82).